|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
85.19.92.6
Hi and sorry for my vague questions ... but i am narrowing and focusing down better my goals. And my present goal are dacs.
The idea is to start from an already decent product and try to improve it with some simple mods.
The products i am interested in are usually very cheap and so some cost cuts have been done.
I wonder how much can be gained in terms of sound from a dac improving its power supply only (unfortunately i have no chance with the digital part).
I would be very interested to get some information/advice from people with experience in this kind of mod.
For instance i have already noticed that even changing the mains transformer only can give some benefits (i guess better filtering of mains noise).
So my feeling is that a lot can be gained even from cheap but already good sounding dacs addressing only their power supplies.
Thanks a lot.
Kind regards,
bg
Edits: 07/19/15 07/19/15Follow Ups:
Many DACs and cdp's can be improved by enhancing the grounding schemes. See the Archives at Hi-Rez Highway for ideas, especially as regards the Sony SCD-1 and 777ES. That might give you a jumping off point for working on other brands and designs.
Hi and thanks a lot for the very valuable advice.
I have a question about " enhancing the grounding schemes "
Does this have the effect of reducing noise ?
Lately i have come to the conclusion that noise is a very bad beast for audio. It tends to cover the details (i.e. low level signals) so important for instance for a good 3D effect.
Another point, but off topic, is the usually poor quality of connectors.
However that can be solved quite easily. And it gives benefits also.
I will search for these mods.
Thanks again.
Kind regards,
bg
Edits: 07/21/15
enhancing the ground scheme should have the effect of reducing noise. With the Sony devices I mentioned, which were built to a cost-no-object level not seen since, the "problem" is that essentially everything is grounded to everything. Unclestu points out that separate power supplies for each element of a CDP is desirable, and the Sony flagship products were built that way. But by redundantly grounding each section to every other section and to the chassis, some of the benefits of separate PSs were lost. Isolating the grounds had a noticeable benefit.
Hi and thanks again
I have indeed a lot to learn before attempting any mod
I find quite surprising that Sony engineers have made some debatable design choices anyway. Not that i doubt the efficacy of the mods.
Actually the contrary.
Thanks again.
Kind regards,
bg
(in audiophile terms) Harman Kardon CD players in the 90's employed extensive ground planing but kept the planes for analog, DAC, Power supply and the board for the focus motors and timing separate.
The planes met at only one single point to minimize cross contamination. Analog stages like many top end multi format players had a shielding box built over the semiconductors and the ground plane on the bottom pof the board.
In addition many players now have extensive shielding across the front faceplate. One of the early top line Wadia machines had a chassis made of two inch billet aluminum with four machined wells to separate the four typical sections of a DAC....
Hi ! thanks again for the very interesting advice.
I have a further question ... after deciding for a single ground on the board would it be beneficial to use pieces of good quality wire (like hook-up wire) to connect that point with different grounds on the board ?
Not very elegant, but effective ?
Thanks again.P.S. sorry for the poor English ...
Kind regards,
bg
Edits: 07/21/15
As I inferred, but did not describe very well, the issue with the Sony CDPs was that "everything was grounded to everything". Plus, the circuit was connected to chassis ground at many different points. The modifications essentially converted the grounding system to "star ground". This is well described in the Hi-Rez Highway thread. As to using "good wire", no audiophile would ever advise another to use "bad wire", so, yes, use good wire. For connection to ground at the star ground point, heavy gauge is de rigeur, copper or silver. Interestingly, once these mods are completed, the circuit works better. The time interval required for loading and reading a CD is reduced, etc.
There must be something wrong! IK understand why this tweak works, how it can be applied, and it would not be hard to implement. What kind of tweak is that?!! Just kidding! Thank you for the very clear explanation.
Dave
Hi actually i do not understand why the designers do not take care of this aspect during the design of the lay-out.
It should be a very well known issue.
Another thing i do not understand.
Sometimes some newer products appear on the market that sound worse than previous models ... this is insane for me.
One step forward and 2 o 3 backwards. Why ?
I remember for instance an amp, the Pioneer a400.
The successor, the a400X, was less convincing soundwise.
Unbelievable ...
Sorry for the outburst.
Thanks again.
Kind regards,
bg
For Sony, the grounding scheme of the original SCD1, etc, may have been a byproduct of their perceived need for modular construction. Thus having the ground for each separate PCB connect to every other PCB ground and to chassis ground requires less thinking on the part of a tech who may be replacing one or another module. But that's just a guess. Surely they knew about star grounding.
Re the fact that new products are not always superior to the products they replace, this happens a lot, all the time, in most markets (cars, audio, washing machines, etc). Some who prefer vintage equipment may view the entire last 40 years in the audio marketplace as a gradual decline in quality. "They don't make them like they used to". Some of this is generated by the need to keep prices under control while also incorporating advances in technology into commercial products. And yet, if you take the long view, I do think "things" we can buy get better over time; it's just not a linear upgoing curve.
HI and thanks a lot for the very helpful reply.
Still when the scope is a TOTL unit excellence in performance should come first and so the design should be not reproachable i guess
Regarding the second point i think that price of parts tends to increase
I do not know a specific case but it could be that they use a weaker transformer ... cheaper parts to keep the price down.
I believe that present TOTL units have very good parts, probably much better than vintage units. But the price is very very high indeed.
I see dacs costing tenths of thousands of USD without any problem.
Thanks a lot again.
Kind regards,
bg
Not to beat a dead horse, but the chips in modern megabuck DACs are certainly "better" by most criteria than the analogous chips of 10 or more years ago (or even last year, in some cases), but the cost differences may often be in the opposite direction; newer technology chips may also be cheaper, or if more expensive, not enough to account even remotely for the differences in pricing of the assembled products. Those stratospheric prices are more aimed at those who have the bucks to play in an exclusive league and who believe at some level that if the item is not super costly, it cannot be good enough for them. This is only my opinion, of course.
Hi and i would think the same if it were not for this multi-bit discussion going around.
Those vintage multi-bit dac chips are object of cult almost.
However i see that TOTL and very expensive newer dacs use sigma-delta chips. They cannot be that bad i guess.
Someone say that there is no comparison with the old dac chips anyway.
Like dacs have reached the top of sound quality in the 90 and then they started to get worse.
For me this is quite difficult to accept.
Thanks again.
Kind regards,
bg
Edits: 07/30/15
since CDPs like the Sony SCD1 and SCD777ES (essentially the same unit but without balanced outputs) still have their cult followers. I always wondered why that was the case, apart from the fact that both ARE built to battleship quality levels almost unheard of these days except at the megabuck level. (In my previous rant I forgot to mention that a large part of the justification for megabuck DACs is based on their elaborate and beautifully made chassis', IMO.) But it's also true that the main chips in the Sony units are superb still by modern standards. The problem is that if those chips fail, the unit becomes about as useful as a doorstop or boat anchor; you'd need an entirely new audio board, because it is near impossible to de-solder and then install just a replacement chip even assuming you could find one. (I get this info from a very skilled professional who worked on my SCD777ES.) I went from a very tweaked SCD777ES to an Oppo BDP105, skipping about 15 years of technological development, yet the Oppo is not superior to the Sony in terms of pure sonics.
Philips in issuing their Crown chipset, the venerable 1541, with only 16 bit 44.1 sampling, claimed why ask for more when all you need is a "perfect" DAC chip for the information at hand?WHile very popular, that did not stop the tidal wave of higher bit rate sets which entered the market, claiming better high frequency filtering because of the higher bit rate.
In essence, DAC's become part of the computer race. Higher numbers as being seen as better. I do not believe the algebraic manipulation of these numbers has improved much though, nor the addressing of certain particular issues as it pertains to music and digital
Edits: 07/31/15
there's no question in my mind that modern cdp's and DACs are far far superior in sonics to the first wave of "perfect sound forever" products, all of which were 16/44. So, something got better. Whether there is continuous linear improvement over time up to the present is quite another matter, however, IMO.
a large part of the issue is in mastering. In analog, its ok to use low level recordings. Not true in digital. You need the bits to get detail and the number of bits is also correlated to volume , unfortunately.
Over the years I notice my discs are getting louder and louder. And they do sound better.
Also, although it receives very little mention, the analog sections are greatly improving. employing judicious filtering to roll off the RFI edginess, although I believe the pendulum has swung a bit too far in that direction.
Apparently some of the sonic improvement has to do with the appreciation of the jitter phenomenon and how to fix it, or so we are led to believe (or beleive).
What was interesting is, as an former HK dealer. I had access to service manuals . In the early 1990's their top line models ( 7500, 7600) had jitter figures in the very low teens ( 12-14), much lower than any separate stand alone DAC and transport. Makes you wonder, and I used all the tricks: reclockers, upwards dithering, trick cables and such.
Hi and thanks a lot again. Very interesting indeed.
I will limit myself only to some caps replacement with better parts and some testing with different transformer.
The risk to do more harm than good is very high for me.
Thanks again for the very valuable advice.
Kind regards,
bg
Hi thanks again and again very very interesting
The importance of a well designed lay-out seems a little underestimated
While i can understand in a cheap mass market product it is more difficult to accept for a flagship model ... and especially from highly famous Brands ...
However ... i will read the thread you mentioned hoping to understand something
I am very slow to catch things and this is one very complex issue indeed.
Amazing the fact that the modded players worked better as well !
Thanks again.
Kind regards,
bg
Edits: 07/22/15
best DACs have separate power supplies for analog, digital,DAC conversion, and thins like the timing circuit.
Each section will cross contaminate each other if a single power supply is used, ot matter how large, and separate smaller supplies is the key with isolation between them
Hi and thanks a lot for the valuable advice.
I think it would be difficult to do that ... but i will keep it in mind.
I have some of this cheaps dacs found on ebay
They usually comes with a cheap wall power adapter.
Just using a better linear supply gives some benefits.
The idea is to extract the board and put some nice and bigger uF in strategic places and use a very low noise linear ps.
First experiments are promising. Noise fights music.
Thanks again.
Kind regards,
bg
Where do you think nose in a DAC is being sourced? The digital bit stream entering the unit contaminates the analog stage and everything else both inside the chassis and including the power supply .
If you have an old TV set, try placing a portable CD player close to the antenna: your picture will immediately show tremendous aberration as the RF will interfere with the reception. Initial CD players used plastic chassis until complaint to the FCC forced them to stipulate that the chassis be made of metal and RFI emission be reduced to certain standards, the main reason why so many early players had optical outputs.
Simply increasing the power supply current does help, but you are not addressing the main issue, at least IMHO
" Where do you think nose in a DAC is being sourced? "
My take is mainly from the power supply but i could be wrong.
Unfortunately i do not have a scope to see how dirty are my mains.
But i have noticed that during sunday the sound is better.
I live in an semi-industrial area.
" The digital bit stream entering the unit contaminates the analog stage and everything else both inside the chassis and including the power supply .
If you have an old TV set, try placing a portable CD player close to the antenna: your picture will immediately show tremendous aberration as the RF will interfere with the reception. Initial CD players used plastic chassis until complaint to the FCC forced them to stipulate that the chassis be made of metal and RFI emission be reduced to certain standards, the main reason why so many early players had optical outputs "
I see. So the digital chips are RFI generators ? this complicates the all issue.
I am using a usb to spdif converter that has isolation transformers on the digital outs (see picture) ... could this be enough ?
" Simply increasing the power supply current does help, but you are not addressing the main issue, at least IMHO "
actually i was focusing more noise than current.
A better PS could suppress more noise coming from the mains and generate less noise by itself. But i understand that i have to study more.
At least i have to try before modding anything.
Thanks a lot again.
Kind regards,
bg
Again in digital equipment noise, in effect, is self generating. You got clock chips, sometimes several, running at RF frequencies. Most are metal covered for some shielding, but I like to ground the cases ( some players and DACs do that, BTW). I believe your unit has three clocks.
The digital processing chips themselves are generally plastic encased and thus RFI can easily enter and exit the chips as they process the bit stream data entering the unit.
All the essential circuits work with RF frequencies and they are not synchronized. They radiate within the chassis and they contaminate all power supplies.
The key IMHO, is to attack the source. Obviously you need the digital data, but it is imperative that you try and contain the "RF radiation" they produce.
Easiest way to limit them is to cover all chips and semiconductors with ERS. There is a reason why they are shielded, and I can see some kind of circuit under neath.
The boxes while acting as a shield are not as effective as ERS, because their reflective surface can actually "bounce" some RF off. Again the lowly HK players used a box but lined it with felt to be more absorbent. You can line the boxes with ERS too.
Some have simply lined the entire chassis with ERS, but that does NOT stop the internal RF reflections from bouncing around and contaminating adjacent chips and circuits. also be aware that many DAC's employ a filter on the analog output stage to roll off RF. I've seen this on Regas and even the Esoteric machines. Using ERS with out disabling the roll off circuit simply makes the unit dull.
I did shielding like I described on an Esoteric unit and it indeed sounded dull till I lifted the small ceramic cap off the output stage. The unit then opened up, with an extremely airy quality and a tremendous sense of depth.
Hi ! thank you very much indeed for the very valuable advice.
I will try some of these EMI/RFI pads for sure ... placing them on the digital chips.
In the meantime i have found a very interesting discussion where you were involved here
http://www.audioasylum.com/cgi/t.mpl?f=tweaks&m=190998
I start studying more seriously the subject.
My main device is now this usb to spdif/aes/optical converter.
I like the fact that gives me the opportunity to link to my pc any dac around and listen.
It is a key device also because in the end it sets the clock.
I have the feeling that is well designed but i have already seen some possible upgrade (in the power supply actually).
Surely addressing the RFI noise it will give further benefit.
I will do the same on the dacs.
Thanks a lot again for the very helpful advice.
Kind regards,
bg
see twisted pear sites
Hi and thanks a lot for the very valuable advice.
I have already some cheap but nice sounding dacs and hope they can be better when fed with a less noisy PS than the stock wall wart.
Thanks again.
Kind regards,
bg
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: